Dashboard instrument cluster repairs

[casey]

Another cluster arrived for repair on Monday afternoon, 2 March. I managed to diagnose and repair and ship it back within 48 hours (it was a pre-booked job).

No surprises, the usual failed IC on the tacho PCB. no needle movement as received.

Tacho removed from cluster and set up on my tacho tester, still dead - but 5 seconds of 100 degrees from a heat gun onto the IC and the tacho sprang into life and worked perfectly. Unfortunately, that "fix" doesn't last long, so I replaced the entire OEM PCB with my replacement design PCB, which uses automotive grade components. All straightforward to install - the hairiest bit is removing the tacho needle with extreme care. Calibration done on my tester, tacho reinstalled in cluster, retested on my cluster test rig and all good.

Pic of tacho with replacement PCB installed:


Video of tacho on tester for calibration;

Pic of cluster on final, full function test:


Two more repairs booked into my diary, but DM me if you have a cluster requiring a repair.

 

[casey]

I have extended my suite of test boxes to enable me to function-test full clusters, plus tachos and CPU boards in isolation, when removed from clusters:

 

[rotors7]

I have extended my suite of test boxes to enable me to function-test full clusters, plus tachos and CPU boards in isolation, when removed from clusters:

View attachment 43558View attachment 43559View attachment 43560View attachment 43561

I'm well impressed with all that !
Awesome

 

[Jester*]

The nerd in me loves this.

 

[casey]

Another cluster turned around in 2 days and currently winging it's way back to the car owner in SWEDEN.

No tacho OR speedo operating, so I was thinking this could be an expensive and time consuming repair. In fact, the cause (which didn't take too long to diagnose) turned out to be a short circuit diode, costing pennies. It was a bit if a faff to remove it and replace with a new one from stock, as the gauge had to be removed from the CPU PCB, but straightforward enough.

 

[casey]

Teaser - 3D CAD image of Speedo daughterboard that can be used to fix "errant speedo needle" issues, when caused by a (now obsolete) failed driver chip. It has taken some time to design and test the electronics, plus a tricky physical design to fit within the space limitations and available fixing points. I've even managed to retain the use of the existing speed calibration variable resistor on the CPU board.

I have just sent off an order for 5 PCB's to be manufactured. Hopefully, they will be perfect, but I'm allowing some time to do a tweak or two when I do some testing.

 

[Jester*]

So cool

 

[RX2761]

Hey I have a faulty tach, I was about to ship it overseas and have just come across this! I'm still relatively new to using the forum, but have been signed up a while. I'll send you a message now!

 

[MILLION]

very nice!

 

[casey]

Hey I have a faulty tach, I was about to ship it overseas and have just come across this! I'm still relatively new to using the forum, but have been signed up a while. I'll send you a message now!

You have a DM

 

[casey]

First batch of "daughterboard" PCB's to replace failed IC6 driver chip on CPU/Speedo board has arrived for assembly, fit and test. I have 3 cluster repairs booked over the next week, so might be a delay in working with these.

 

[casey]

Another cluster will be winging it's way back to the owner tomorrow. This was a weird fault/repair. Customer reported intermittent tacho, sometimes stuck at zero, other times OK, initially pointing to an intermittent electronic issue, maybe a dry joint or component failure.

The tacho, when tested in the cluster on receipt was, indeed, stuck on zero. When removed from the cluster and tested on it's own, it worked perfectly. The more I thought about it, the more it seemed like a mechanical fault, requiring more "energy" to lift the needle off zero. Bingo, that was it. I have a very close look at the needle stop on the gauge face and noted a physical "gouge" in the stop pin and a corresponding faint mark on the needle.

This video shows the needle stuck at zero, but a slight nudge would get it moving. This was reproducible. You can even hear a slight click as the needle, hooked onto that gouge, breaks free.

The solution was to place a heat shrink sleeve over the damage to stop the sticking, and clean the mark on the needle. You'd never know the sleeve was there.

A very unique failure mode!

 

[casey]

Right, two more clusters arrived today for repair, so the fun continues.....